Related Experiment Video
Updated: Apr 24, 2026
12:18
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
7.0K
Impact of the Interaction Between Screen Time and Activity Interests on Adolescent Depression Risk: Construction of a
Leiming Mao1, Ruiqi He2, Shike Zhang2
1Nantong Mental Health Center, 226001 Nantong, Jiangsu, China.
Actas Espanolas De Psiquiatria
|April 23, 2026
Summary
Machine learning models can identify adolescents at risk for depression by analyzing screen time and activity interests. Increased activity interests protect against depression, even with high screen exposure.
Area of Science:
- Adolescent mental health research
- Machine learning applications in public health
- Behavioral science
Background:
- Adolescent depression is a growing public health concern.
- Excessive screen time is linked to increased depression risk.
- Activity interests may offer protective benefits against depression.
Purpose of the Study:
- To develop a predictive machine learning (ML) model for adolescent depression risk.
- To evaluate the combined effects of screen time and activity interests on adolescent mental health.
- To explore novel analytical methods beyond traditional approaches.
Main Methods:
- A multi-center survey of 2202 adolescents aged 10-14 years.
- Utilized the Child and Adolescent Mental Health Screening Questionnaire with seven validated scales.
- Employed machine learning models (logistic regression, XGBoost, CatBoost) with feature selection and class imbalance techniques.
Main Results:
- Activity interest and psychological functioning were key predictors of depression risk.
- Machine learning models revealed nonlinear screen-time effects and dose-dependent protective effects of activity.
- CatBoost demonstrated balanced performance, while logistic regression showed better generalizability.
Conclusions:
- Machine learning models can effectively screen adolescents for depression risk.
- The models integrate screen time and activity interests for comprehensive assessment.
- Findings support school-based early identification of at-risk youth.
